While trying to make a fresh install with the from the Oct 12th miniroot58.fs,
while installing the sets I get a bunch of:
tar: Unable to set file uid/gid of ./blah/blah: Operation not permitted
tar: Unable to set file uid/gid of ./...: Operation not permitted
tar: Unable to set file uid/gid of
On Thu, Jun 25, 2015 at 07:01:29AM +, Pavel Plamenov wrote:
There are some leftover links to www@, which is gone.
I sent an almost identical diff over a year ago:
https://www.marc.info/?l=openbsd-miscm=139627200904849w=2
I think this is the right direction.
SIGINFO is awesome, but it's even better when it actually does
something relevant.
This makes it print the total counts so far to stderr.
Useful? Feature creep? You decide.
Index: wc.c
===
RCS file: /cvs/src/usr.bin/wc/wc.c,v
I now realize this may have been ignored simply because the clock
on the sending machine was horribly off and many people sort mail
by date. So... Should this go in? Am I missing something?
On Thu, Apr 30, 2015 at 06:03:23PM -0400, Jean-Philippe Ouellet wrote:
The intermediate values calculated
On Sat, Jun 20, 2015 at 03:01:18PM +0200, Reyk Floeter wrote:
there is some great interest in getting support for rewrites
What do people think of something like our tftpd(8)'s -r
-r socket
Issue filename rewrite requests to the specified UNIX domain
socket. tftpd will
This adds a -p option to pkg_info to show the PackageRepositorys
being used.
Inspired by trying to parse /etc/pkg.conf with some awk and quickly
realized that was the wrong way to solve my problem. I then wrote
some perl that reached into OpenBSD:: internals, but concluded it'd
be much cleaner
This is no longer used anywhere, and even crashes if you try to call it.
Besides, OpenBSD::PackageRepository::Source was removed almost a year ago.
Index: PackageRepositoryList.pm
===
RCS file:
The idea was to print the package repository used followed by the stuff
found in it, but that doesn't work:
$ pkg_info -vQ foo
Use of uninitialized value $ENV{PKG_PATH} in concatenation (.) or string at
/usr/libdata/perl5/OpenBSD/PkgInfo.pm line 604.
PKG_PATH=
foo2zjs-20140627
The overhead is somewhat high, and it's considered broken anyway:
https://www.acsac.org/2012/openconf/modules/request.php?module=oc_proceedingsaction=view.phpa=Acceptid=237type=4
P.S. Sorry for breaking threading, my mail setup is currently a mess.
The intermediate values calculated in hmac_sha1 as part of
pkcs5_pbkdf2 are not zeroed afterwards, so we leak a single-hashed
version of the key on the stack in tk[].
Also, the correct RFC defining this is
RFC 2104 - HMAC: Keyed-Hashing for Message Authentication
not
RFC 2202 - Test Cases
POSIX says the truncated things are record(s) not block(s):
http://pubs.opengroup.org/onlinepubs/9699919799/utilities/dd.html
That's what it's historically been too:
http://minnie.tuhs.org/cgi-bin/utree.pl?file=V7/usr/src/cmd/dd.c
We use async-signal-safe (not async-signal safe) elsewhere,
and so does POSIX.
http://pubs.opengroup.org/onlinepubs/9699919799/functions/sigaction.html
Index: sigaction.2
===
RCS file: /cvs/src/lib/libc/sys/sigaction.2,v
retrieving
, I just made sure it still built and that build built itself
too.
Capsicum is still blocking on this patch (or one like it).
Any/all feedback welcome.
On Thu, Jul 10, 2014 at 04:13:38PM -0400, Jean-Philippe Ouellet wrote:
This diff adds another struct between filedesc and file to store
process
Ping?
On Sun, Jul 13, 2014 at 03:45:44PM -0400, Jean-Philippe Ouellet wrote:
Updated for mallocarray() and free(size).
On Thu, Jul 10, 2014 at 04:13:38PM -0400, Jean-Philippe Ouellet wrote:
This diff adds another struct between filedesc and file to store
process-local per-descriptor
On Mon, Jul 21, 2014 at 06:59:12AM +, Doug Hogan wrote:
-objects and checks for arithmetic overflow.
+objects and calls
+.Xr panic 9
+on arithmetic overflow.
That is misleading in the M_CANFAIL case.
I'm not terribly good at wording things, but I suggest something
more like this
For the cases where it's more than just nitems * sizeof(item),
maybe it wouldn't be a bad idea to have something like:
static __inline int
MULT_OVERFLOWS(int x, int y)
{
const intmax_t max = 1UL sizeof(size_t) * 4;
return ((x = max || y = max) x 0 SIZE_MAX / x y);
}
(or
The C standard mandates that static be first.
From ISO/IEC 9899:1999 and 9899:201x,
6.11.5 - Storage-class specifiers:
The placement of a storage-class specifier other than at the
beginning of the declaration specifiers in a declaration is
an obsolescent feature.
and -Wextra
On Mon, Jul 14, 2014 at 11:44:30AM -0400, Jean-Philippe Ouellet wrote:
The C standard mandates that static be first.
Of course I forgot something... This is the hunk that made me
notice in the first place. Found while porting signify to osx.
Index: lib/libc/hash/sha2.c
On Sun, Jul 13, 2014 at 04:03:53PM +0200, Brent Cook wrote:
On Jul 13, 2014, at 3:58 PM, Ted Unangst t...@tedunangst.com wrote:
@@ -411,6 +404,9 @@ static long
random_l(void)
{
int32_t i;
+
+ if (use_arc4random)
+ return arc4random() 0x7fff;
return
On Sun, Jul 13, 2014 at 11:29:22AM -0600, dera...@cvs.openbsd.org wrote:
- ldp = malloc(sizeof(*ldp) + (k-1), M_DEVBUF, M_NOWAIT);
+ ldp = mallocarray(k-1, sizeof(*ldp), M_DEVBUF, M_NOWAIT);
Are you sure k-1 can never be small enough such that a*b is less than a+b?
And some cosmetic things:
Index: ic/malo.c
- ring-data = malloc(count * sizeof (struct malo_rx_data), M_DEVBUF,
- M_NOWAIT);
+ ring-data = mallocarray(count, sizeof (struct malo_rx_data),
+ M_DEVBUF, M_NOWAIT);
Might as well s/sizeof (/sizeof(/ while you're here.
Updated for mallocarray() and free(size).
On Thu, Jul 10, 2014 at 04:13:38PM -0400, Jean-Philippe Ouellet wrote:
This diff adds another struct between filedesc and file to store
process-local per-descriptor information. Currently, the only thing in
this struct is the file pointer and some
It takes a void *, not a char *.
Index: getentropy.2
===
RCS file: /cvs/src/lib/libc/sys/getentropy.2,v
retrieving revision 1.4
diff -u -p -r1.4 getentropy.2
--- getentropy.215 Jun 2014 07:24:19 - 1.4
+++
On Thu, Jul 10, 2014 at 12:02:40PM -0700, Matthew Dempsky wrote:
-.Fn malloc unsigned long size int type int flags
+.Fn malloc size_t size int type int flags
+.Ft void *
+.Fn malloc size_t nmemb size_t size int type int flags
2nd one should be mallocarray.
This diff adds another struct between filedesc and file to store
process-local per-descriptor information. Currently, the only thing in
this struct is the file pointer and some flags, however I have another
patch on top of this that adds capsicum capabilities to it. (And
another that uses
Hi,
When assigning mount points to an already-partitioned disk
without a its fstab handy, it's annoying to type the partition
letters one after another. (And sing the alphabet each time or
look at the output of 'p' for letters after 'f'. I blame hex.)
I could go back to preeschool and learn the
Eww...
See distrib/notes/mirrors and installpath from pkg.conf(5).
On Wed, Jun 04, 2014 at 08:02:06PM +, Miod Vallat wrote:
First, str{cat,cpy} were vehemently expunged from the kernel many years ago,
so stop trying to keep them around.
Index: lib/libc/Makefile.inc
Hello, this is libc you are butchering in. I'm afraid strcat and strcpy
are still
Hello,
This came up when I was looking for the proper place to put code for dealing
with capsicum data structures which need to be handled by both userland and
the kernel.
FreeBSD's libc build system has tentacles that reach over and grab
sys/kern/subr_capability.c. That's not very elegant, I
On 26/05/14(Mon) 13:46, Martin Pieuchot wrote:
I'd appreciate if people having troubles with suspend/resume could try
this diff an report back.
Fixes it for me! :D
Many thanks.
Previous diff was lacking the header chunk, please use this one instead.
Was the corresponding commit missing the
Found while trying to figure out how to be sure the console is always
cleared at logout (and the resulting login prompt seemed like a pretty
guaranteed-to-always-work place). If you have a cleaner way, perhaps
one that is actually at logout and doesn't involve missing all the
startup messages due
todd, can you put this in snaps so that we know if there's some fallout?
I don't know if this is related or not, but running that snap I experienced
the following crash while trying to shut down: (which I've never had before)
http://i.imgur.com/4YFzdv0.jpg
http://i.imgur.com/KR1hWtT.jpg
There are a bunch of nearby printfs which start with : ... without
a devname. I'm not sure which one is preferred, so here's both ways.
Index: ehci_cardbus.c
===
RCS file: /cvs/src/sys/dev/cardbus/ehci_cardbus.c,v
retrieving
On Mon, May 05, 2014 at 07:31:34PM +1000, Joel Sing wrote:
This one is calloc, not reallocarray, so unless I'm seriously missing
something obvious here, it is indeed zero'd, no?
Run the following before and after your change:
Ah, yep. Can't believe I missed that (along with all the other
On Sun, May 04, 2014 at 12:17:16PM -0600, Theo de Raadt wrote:
We are going to completely ignore diffs which change multiple idioms
at once.
Okay.
That is how mistakes get made.
Yep, more true than I realized.
Here's a simpler one:
Index: apps.c
On Sun, May 04, 2014 at 11:30:40PM +0200, Alexander Hall wrote:
NULL theoretically could be != 0
Umm... short of something like:
#undef NULL
#define NULL I'm silly and want to break everything
or something, I don't see when that'd be the case.
According to ISO/IEC 9899:1999 TC3 (n1256)
On Mon, May 05, 2014 at 11:12:00AM +1000, Joel Sing wrote:
- i = 0;
if (arg-count == 0) {
arg-count = 20;
- arg-data = (char **)malloc(sizeof(char *) * arg-count);
+ arg-data = calloc(arg-count, sizeof(char *));
}
- for (i = 0; i
Hello,
When building libcrypto on amd64 I get this warning:
(cd /usr/src/lib/libcrypto/crypto/../../libssl/src/crypto/md5 ; /usr/bin/perl
./asm/md5-x86_64.pl openbsd-elf) md5-x86_64.S
Use of uninitialized value $output in pattern match (m//) at
./asm/md5-x86_64.pl line 115.
Admittedly it's
This doesn't fix the problems, only removes markers alerting us to audit it.
Memory management in these files is still missing integer overflow checks,
NULL return checks, and is full of crazy abominations like:
X509_NAME *
parse_name(char *subject, long chtype, int multirdn)
{
size_t
Hi,
Here's more fuel for the OpenSSL fire. Mostly just axeing at ifdefs,
trying to err on the conservitive side.
There's obviously *TONS* more to clean up, but I only had so much time
tonight. :)
BTW, libssl and libcrypto don't currently build because their Makefiles
still include some
Thank you very much for the feedback.
On 3/14/14 9:38 AM, Ingo Schwarze wrote:
According to the sigaction(3) manual, volatile sig_atomic_t would
be better. If i understand correctly, overzealous compilers might
otherwise optimize checks away.
Dammit, of course. I should have caught that.
On 3/12/14 11:15 PM, Loganaden Velvindron wrote:
I've read about the file vulnerability, and capsicumization also
came to mind. However, there was also a discussion when i was
playing with capsicum and openssh, about the limits of capsicum.
Capsicum doesn't prevent DoS, and we still need
On 3/13/14 2:39 AM, Loganaden Velvindron wrote:
I'm not a mentor, but I'd be happy to help you in any way I can.
You can send mails to tech@ for testing your diffs.
Any chance you'd like to review my bootloader patch from last month then?
http://marc.info/?l=openbsd-techm=139408992902933
I
On 3/13/14 3:18 AM, Loganaden Velvindron wrote:
On 3/13/14 10:57 AM, Jean-Philippe Ouellet wrote:
On 3/13/14 2:39 AM, Loganaden Velvindron wrote:
I'm not a mentor, but I'd be happy to help you in any way I can.
You can send mails to tech@ for testing your diffs.
Any chance you'd like
On 3/12/14 4:58 AM, tuchalia wrote:
Hi all,
I'm really interested in this possibility of porting the Capsicum framework
to OpenBSD. Should l try to port also the Casper daemon to OpenBSD, or
only work in the kernel implementation?
I've used Capsicum during the last summer, but I only
On 3/12/14 4:58 AM, tuchalia wrote:
Also, do we have any IRC channel to discuss al this?
I've been wondering about that too, although I was never really active
on any of the channels.
Mindcry is dead, subcult is mostly non-english, freenode and efnet are
mostly whiners. I vaguely remember
Hello,
When lock(1) receives SIGINT, SIGQUIT, or SIGTSTP, it calls hi()
twice, once because it's the signal handler, and once after
readpassphrase() errors because the read was interrupted.
Since hi() gets called when readpassphrase() fails anyway, this
patch ignores the signals instead of using
On 3/12/14 4:58 AM, tuchalia wrote:
Should l try to port also the Casper daemon to OpenBSD, or
only work in the kernel implementation?
Based on more private mail, I figured it'd be a good idea to make what I
plan to work on public in case there are others interested so we can
avoid stepping on
On Wed, Mar 12, 2014 at 11:09:14PM +0100, Ingo Schwarze wrote:
I don't really like the warnx(3) call from the bye() ALRM handler
either, but that's a separate matter.
Me neither.
Maybe something like this instead? (although maybe the done check should
be someplace else?)
Index: lock.c
do
it, but I'm not sure what the cleanest approach would be. Any feedback
is much appreciated.
Original message with patch:
http://marc.info/?l=openbsd-techm=139208386213686q=raw
On 2/10/14 8:57 PM, Jean-Philippe Ouellet wrote:
Hello,
I patched the sparc64 bootloader to allow users
Hello,
I patched the sparc64 bootloader to allow users to manually specify
network config and where to load the kernel from via openfirmware
parameters instead of always requiring rarp/bootparams/bootp.
This enables remote bootstrapping of semi-recent sun boxes (like
the T1000) on networks where
Hello,
While hacking on the sparc64 bootloader (patch in another mail) I
had some questions about the recently added super-early entropy
loading code.
http://www.openbsd.org/cgi-bin/cvsweb/src/sys/arch/sparc64/stand/ofwboot/boot.c.diff?r1=1.19;r2=1.20
From the commit message:
Try to load
Here's some documentation of it for www/
I think there should also be some mention of this functionality in
diskless(8), but I don't know where to put it, and I don't want to
just .Xr it to boot_sparc64 because diskless is for all archs and
none of the other boot_*s are referenced.
Index:
It appeared in revision 1.3 (Update from lite2.)
It's the only one in the string family that has it, and nothing from it
is used.
Index: strsep.c
===
RCS file: /cvs/src/lib/libc/string/strsep.c,v
retrieving revision 1.6
diff -u -p
A few days ago I saw a commit for NSD, I had never heard of it before,
so naturally, I went to read the manpage, however it wasn't there. I
looked at the cvs tree, and saw that there was a manpage, just not
formatted for mandoc like all other manpages I've seen in OpenBSD, so I
read up on
55 matches
Mail list logo